Power of grammatical evolution neural networks to detect gene-gene interactions in the presence of error

BMC Research Notes

Table 5 Power of MDR (from Ritchie et al. 2007) to detect either genetic model in data with genetic heterogeneity.

Source of Noise	Power (%) to Detect Either Model (5,10 or 3,4)
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
GH	70	34	42	41	20	19
GE+GH	69	34	42	41	20	19
GH+PC	24	35	9	8	7	5
GH+MS	65	40	42	31	18	22
GE+GH+PC	27	35	10	8	7	6
GE+GH+MS	64	44	42	41	16	11
GH+PC+MS	23	38	9	10	4	6
GE+GH+PC+MS	31	36	9	7	4	3

GE = 5% Genotyping Error; GH = 50% Genetic Heterogeneity; PC = 50% Phenocopy; MS = 5% Missing Data

ISSN: 1756-0500